AbstractInadequate vitamin D levels have been associated with increased risk of depression. However, most of these studies are cross-sectional and failed to investigate the effect of changes in vitamin D levels. This study aimed to investigate the longitudinal association of changes in serum 25-hydroxyvitamin D levels with depressive symptoms in 3365 participants of the English Longitudinal Study of Ageing, a large nationally-representative study of older adults. Based on their vitamin D levels at baseline and follow-up (sufficient ≥ 50 nmol/L; insufficient < 50 nmol/L), participants were classified as follows: with sufficient levels at both waves; with sufficient levels at baseline but not at follow-up; with insufficient levels at baseline but ≥ 50 nmol/L at follow-up; and with levels < 50 nmol/L at each time point. Depressive symptoms were measured using the 8-point CES-D scale. Data were analysed using logistic regression models. Compared with those with sufficient levels of vitamin D at both waves, only those with insufficient levels throughout were more likely to report elevated depressive symptoms (OR = 1.39, 95% CI = 1.00–1.93). Becoming or no longer being vitamin D deficient was, in the short term, not associated with elevated depressive symptoms. Further evidence is required on whether vitamin D supplementation might contribute to the prevention or treatment of depression as well as on the duration of time for changes in vitamin D levels to lead to detectable changes in depressive symptoms.